On Friday, a Florida man was arrested in connection with a string of mail bombs sent to prominent critics of President Trump, including Barack Obama, Hillary Clinton, and Kamala Harris. Reporters spent the day working to unearth the social media accounts of the man, 56-year-old Cesar Sayoc, Jr. of southern Florida. While the authenticity of every account has yet to be confirmed, reporting so far paints a picture of a man who bought into many of the right-wing conspiracy theories that burbled up from social media, to mainstream conservative media, to the president himself.
Among other things, Sayoc appears to have been an active Trump supporter on Facebook, and an enthusiastic troll on Twitter. Here’s Adi Robertson:
The account @hardrock2016, which appears to belong to Sayoc, sent veiled death threats to liberal “slime scum,” including Vice President Joe Biden, former New York attorney general candidate Zephyr Teachout; actor Jim Carrey; Portland, Oregon mayor Ted Wheeler; MIT Media Lab founder Nicholas Negroponte; and New York Times journalist (and former Verge writer) Sarah Jeong.
At least one of the targets — Biden, who was threatened on September 18th — was among those who later had bombs addressed to them. At least one recipient — former congressional press secretary Rochelle Ritchie — apparently reported the threats to Twitter, but was told that they didn’t violate Twitter’s abuse rules. “This is an ongoing law enforcement investigation. We do not have a comment,” a Twitter spokesperson told The Verge when asked whether the messages complied with its guidelines.
One researcher, Jonathan Albright, counted the number of times that Sayoc replied on Twitter to celebrities and figures with a meme about a Parkland shooting survivor being a “crisis actor” paid by George Soros: 59.
It’s a disturbing irony. Because as authorities continued to discover new mail-bomb targets on Friday, the right wing flooded media channels with suggestions that the bombs themselves were part of a Democratic hoax. It’s a toxic circle: Man falsely enraged by crisis actors allegedly sends bombs; conservative media falsely describes bomb targets as crisis actors.
In a Twitter thread, Albright chronicled how conservatives were able to reach a much wider audience with their hoax claims on Instagram, using various features of the platform. The right wing adopted the hashtag #Soros to share many of these memes, and Instagram helpfully organized the most-engaged posts algorithmically. It auto-populated suggested searches for anyone who began to search for Soros: “soros caravan,” “soros bomb,” “soros jew,” all of which could lead users to further misinformation.
Instagram search results also auto-populated with a bunch of obviously fake Soros accounts, although many of them appear to have been taken down overnight.
On Twitter, a similar phenomenon played out, as Blake Montgomery charted at BuzzFeed. Hashtags, as usual, raced ahead of the truth:
But people on Twitter, including right-wing commentators with name recognition like Ann Coulter, James Woods, and Candace Owens, tweeted that the devices, described as being similar to pipe bombs, were a scheme concocted by Democrats to boost sympathy and turnout before the midterm elections in November. However, there is no evidence to support their claims. And neither the identity nor political affiliation of the perpetrators are known.
Still, #FAKEBOMBSCARE, #FakeBombs, and #FalseFlag — all dedicated to the conspiracy theory — trended alongside #BombScare. Many used #BombScare to tweet the theory as well, but the hashtag itself is not blatantly false like the others. #MAGABOMBER, a hashtag devoted to the idea that the bomber was a right-winger attacking the president’s nemeses, also trended, again with no proof.
In part, this is a now-old story about how social media spreads misinformation in the immediate wake of the crisis. But if Sayoc is indeed the bomber, and these social media accounts belong to him, it suggests something even more disturbing: a person steeped in conservative media, radicalized into violent action — at the same time the same echo chamber, all evidence to the contrary, dismisses a series of attempted assassinations as a hoax.
The platforms have their part to play in reducing the polarization that now consumes us. But as Albright wrote earlier in the week, in a piece about how a false meme spread alleging that Soros had funded the caravan of refugees coming to America, the infrastructure that promotes this misinformation is quite powerful. Whatever captures our attention, if we simply stare at it long enough, becomes real.
”The legitimization of the narrative happens through attention,” he wrote. “It is reinforced in click-seeking MSM headlines and fact-checks through keyword repetition. It is platformless.”
Facebook’s said today that it had removed a combined 82 pages, groups, and accounts that were masquerading as US and sometimes UK citizens and organizations. In fact, they were part of an influence campaign linked to Iran. (For many examples from these campaigns, check out the Digital Forensics Research Lab’s post.)
Facebook says it removed 30 pages, 33 Facebook accounts, and three groups on Facebook, and it found 16 new accounts on Instagram. The accounts and pages spent less than $100 on advertising, and they only hosted or co-hosted a total of seven events. However, about 1 million people followed at least one of the pages, while roughly 25,000 people joined at least one of the groups. On Instagram, around 28,000 people followed at least one of the Iran-linked accounts. The earliest account was created in June of 2016, but they were most active over the last year, Gleicher told reporters on a press call this morning.
Paulo Trevisani and Deepa Seetharaman examine the state of social media and democracy in Brazil ahead of this weekend’s presidential election:
Earlier this week, a video shared widely on social media showed a man resembling a leading gubernatorial candidate in São Paulo state naked in an orgy. The candidate, João Dória, a married father of three, denies it was him in the video and is asking electoral authorities to find its creator.
“We don’t have a ready, efficient solution for fake news,” said Brazil’s top electoral authority, Judge Rosa Weber, to reporters earlier this week, as investigators fought back internet rumors that Brazil’s electronic voting stations had been tampered with. “We haven’t found a miracle.”
The president is concerned that his follower count is going down as Twitter purges itself of bots.
Television is still king when it comes to political advertising reports Karl Evers-Hillstrom:
Digital ad spending is reaching new heights during the 2018 election cycle, but in tight, contested congressional races, TV advertising is still king.
In total, Senate candidates have spent nearly $150 million on TV ads since May 31, compared to $17.5 million on Google and Facebook ads, according to a Wesleyan Media Project report — produced in partnership with the Center for Responsive Politics — released last week.
Kevin Collier reports on how states are communicating disinformation about voting to the big platforms:
For Facebook, which began the process in October, it’s straightforward. Staff at secretaries of states’ and state election directors’ offices have been given a particular email address that forwards to a dedicated Facebook staff to review and potentially remove it.
With Twitter, it’s a little more complicated. Two organizations, NASS and the National Association of State Election Directors, its sister organization, have been given trusted status on the company’s election issues portal, created this summer. If Twitter staff decide an account or string of accounts is part of a larger effort to deceive voters, it will delete them all.
The UK’s data watchdog, the ICO, gave Facebook the maximum possible fine over its failure to protect user data in the Cambridge Analytica scandal: £500,000. It would have been much higher if GDPR had been in effect.
Rosemary Ajayi writes about a series of incidents in which people impersonated Nigerian politicians in an effort to scam them out of money:
Impersonation is a growing problem on Nigerian Twitter with some accounts set up using politicians’ names to promote propaganda while others snag celebrity names as part of marketing fraud schemes. Often, the differences are slight and difficult to detect for the unsuspecting user.
Earlier this year, Femi Otedola, a billionaire Nigerian oil tycoon, publicly quit Twitter after the company failed to remove his impersonators. He used the platform for only six weeks. His Twitter handle has now been verified, although he is yet to resume tweeting.
You’d think Facebook would get a discount on this! But no:
The social network has spent almost $12.5 million in the past six months to promote its own election integrity products, including the ad archive itself, which shows users who paid for what political ads. (Even though Facebook owns the platform, it still runs ads through the same system as other advertisers. The money simply changes pockets.)
Facebook’s top political spender besides Facebook is the Democratic candidate for Senate, reports April Glaser:
One particular tactic that’s set the Beto for Texas campaign apart it is reliance on ads on Facebook, where it’s shelled out more than $5.73 million since May, according to an ad-spending report released by Facebook on Tuesday. The only other candidate running for a seat this November who comes somewhat close is billionaire businessman J.B. Pritzker, a Democrat who has spent nearly $2 million on Facebook in his bid for Illinois governor. After Pritzker, the Facebook figures drop dramatically: The next biggest ad buyers running in November are all Democrats spending less than $1 million each: Sen. Claire McCaskill of Missouri, Sen. Heidi Heitkamp of North Dakota, and Sen. Kirsten Gillibrand of New York.
Counterpoint: asking big tech to not police tech speech has also proven to be a dangerous path! But here’s some more from Alex Stamos:
STAMOS: The next step we need is federal legislation to put a limit on ad targeting. There are thousands of companies in the internet advertising ecosystem. Facebook, Google, and Twitter are the only ones that have done anything, because they have gotten the most press coverage and the most pressure from politicians. So without legislation we’re just going to push all of the attackers into the long tail of advertising, to companies that don’t have dedicated teams looking for Russian disinformation groups.
Rough earnings report for Twitter on Thursday:
Twitter’s monthly user count took another hit last quarter, diving by 9 million amid an ongoing crackdown meant to rid the platform of bots, spam, and other problematic accounts. The company said this morning that its monthly user base fell to 326 million from from 335 million the prior quarter, marking the second quarter in a row of declines.
Snap also had a rough earnings report, losing 2 million users over the previous quarter.
Josh Constine reports that Facebook is building a murder-clone of Bytedance’s TikTok called Lasso, hoping to capitalize on teens’ enthusiasm for recording short lip-synch videos. And no wonder: ByteDance, which makes TikTok, was just valued at $75 billion.
Here’s an overdue profile of the great Jane Manchun Wong, who digs into source code to see what companies like Facebook are testing in their apps before it’s ever made public.
Wong is currently still a student, but when she graduates she hopes to find a job at one of the platforms she reverse-engineers. For now, her work is for fun: “It feels like going on an adventure, like treasure hunting… To me, my reward is to be able to see what’s new, what’s coming next.”
I wrote about Snap Camera, a new desktop app that lets you stream yourself wearing thousands of Snap lenses on apps including Twitch, YouTube, and Skype. It’s cute!
Feature request: a sticker pack about not murdering the people who live in your village.
Ahead of the launch of Lasso, Facebook is adding a new ‘music’ section on profiles, bringing music stickers to Stories, and rolling out its Lip Sync Live game to pages.
Oculus is rolling out new updates that include a platform-wide abuse reporting system, Adi Robertson reports It’s live on the Samsung Gear VR and Oculus Go, and is coming to the Oculus Rift later.
Facebook is enticing political candidates to make vertical videos in which they deliver monologues to voters.
Max Read considers why Nick Clegg could be a good fit for Facebook, where he is taking over communications and policy:
But while the left-liberal vision of an “open society” has increasingly failed to move voters over the last decade, over that same period of time it’s been an animating principle of one of the world’s largest and most powerful companies. Facebook may be used around the world to stoke nationalism and populist anger, but its ostensible guiding values are remarkably similar to those of the Liberal Democrats. One imagines that Clegg, a longtime proponent of the open society — with its premium on “the sharing of knowledge and information,” its “internationalist outlook,” and its belief that “all are free to rise” — might have looked at Facebook and seen a group of ideological fellow travelers. More to the point, doesn’t it seem likely that Clegg, examining the dire state of the Liberal Democrats in the U.K., might see in Facebook a better political partner for the pursuit of his goals? In that sense Clegg isn’t leaving politics for tech so much as exchanging one form of politics — the ballot box — for another — the platform. If you can’t win at the polls, win in the app store, maybe.
Ben Thompson writes about why Facebook’s Oculus division may have a limited upside:
In other words, the virtual reality market is fundamentally constrained by its very nature: because it is about the temporary exit from real life, not the addition to it, there simply isn’t nearly as much room for virtual reality as there is for any number of other tech products.
And finally ...
Usually when we talk about 4chan, it’s in the context of users creating hateful memes and directing them a wide variety of targets. But here is a story about one of them maybe solving a prominent problem in mathematics. It all started on a thread about the shortest sequence of episodes you would need to watch if you wanted to watch one season of a beloved anime show named Haruhi in every possible order. An anonymous 4chan poster posted an elegant solution that has since led to further breakthroughs in the field:
Beyond answering obscure anime questions, there are no known applications for the formula, which isn’t unusual in the field. It often takes decades, Pantone says, for formulas that are discovered in pure mathematics to make their way into real-world applications. But the 4chan episode does show that math can be accessible to anyone.
“This proof shows that you don’t need to be a professional mathematician to understand mathematics and advance the frontier of knowledge,” Pantone says. “That’s the beautiful thing about math, is that anyone can understand the questions.”
Anyone can understand the questions — except for maybe me. Anyway, thanks 4chan!
Talk to me
Send me tips, comments, questions, and weekend plans: firstname.lastname@example.org.